Fig. 8: ST2⁺ T_reg cells kill intratumoral (TME) CD8 T cells in the leukemic niche.

A In vitro direct killing assay of TME BM CD8 T cells by TME BM ST2⁺ T_reg cells. Coculture of AML-activated BM ST2⁺ T_reg cells or non-activated BM ST2⁻ T_reg cells with SYTOX-Blue viability dye-stained TME BM CD8 T cells at a ratio of 1:1, treated with or without IL-33 (20 ng/ml) or GZMB inhibitor Z-AAD-CMK (50 ng/ml). Data are mean ± s.e.m. (n = 4); an unpaired two-sided t-test was used. B In vitro direct killing versus through transwell assays. Coculture or transwell of AML-activated BM ST2⁺ T_reg cells or non-activated BM ST2⁻ T_reg cells with SYTOX-Blue viability dye-stained TME BM CD8 T cells at ratios ST2⁺ T_reg cells: CD8 of 4:1, 2:1, and 1:1. Data are mean ± s.e.m. (n = 4); unpaired two-sided t-test was used. C ST2⁺ T_reg cells do not directly kill leukemic cells, myeloid cells, or B cells. Killing ability of TME-derived BM ST2⁺ T_reg cells in coculture at the indicated ratios with MLL-AF9^eGFP leukemic cells, myeloid cells (CD45⁺CD11b⁺), and B cells (B220⁺). Data are mean ± s.e.m. (n = 4); an unpaired two-sided t-test was used. D Ex vivo direct killing assay of TME BM CD8 T cells by TME BM ST2⁺ T_reg cells. Representative flow images of Foxp3^eGFP (Green) T_reg cells, anti-CD8-PE CD8 T cells (Red), and SYTOX-Blue AML cells (Purple) immunofluorescence in the cocultures of TME BM-derived CD8 T cells and TME BM-derived T_reg cells at the indicated ratios, treated with or without IL-33 (20 ng/ml) or GZMB inhibitor Z-AAD-CMK (50 ng/ml). Statistically analyzed lysis percentages of CD8 T cells killed by surrounding T_reg cells (SYTOX-Blue⁺ PE⁺ eGFP⁺) from each coculture group (n = 4). Data are mean ± s.e.m.; an unpaired two-sided t-test was used. E Coculture of TME BM ST2⁺ T_reg cells with CD45RA-sorted naïve CD8 T cells, non-TME-derived BM CD8 T cells, or TME-derived BM CD8 T cells from the MLL-AF9 leukemia model, at the indicated ratios and comparison of the cytolytic effect of these ST2⁺ T_reg cells on these different CD8 T cells (n = 4). Naïve CD8 T cells were sorted for the CD45RA marker from naïve mice; non-TME CD8 T cells were sorted from naïve mice; TME CD8 T cells were sorted from the MLL-AF9 leukemic mice on day 10. These three different CD8 T cell populations were used in killing assays. Data are mean ± s.e.m.; an unpaired two-sided t-test was used. F Splenic ST2⁺ T_reg cells kill leukemic spleen-derived CD8⁺T cells in the MLL-AF9 model. Coculture of TME Spleen (Sp) ST2⁺ T_reg cells with CD45RA-sorted naïve CD8 T cells or non-TME-derived Sp CD8 T cells, or TME-derived Sp CD8 T cells from the MLL-AF9 leukemia model, at the indicated ratios and comparison of the cytolytic effect of these Sp ST2⁺ T_reg cells on these different CD8 T cells (n = 4). Naïve CD8 T cells were sorted for the CD45RA marker from naïve mice; non-TME CD8 T cells were sorted from naïve mice; TME CD8 T cells were sorted from the MLL-AF9 leukemic mice on day 10. These three different CD8 T cell populations were used in killing assays. Data are mean ± s.e.m.; an unpaired two-sided t-test was used. G ST2⁺ T_reg cells kill DNMT3A/FLT3^ITD leukemic BM-derived CD8⁺T cells. Coculture of TME BM ST2⁺ T_reg cells with non-TME-derived BM CD8 T cells, or TME-derived Spleen CD8 T cells from the DNMT3A/FLT3^ITD leukemia model, at the indicated ratios and comparison of the cytolytic effect of these ST2⁺ T_reg cells on these different CD8 T cells (n = 4). Non-TME CD8 T cells were sorted from naïve mice; TME CD8 T cells were sorted from the DNMT3A/FLT3^ITD leukemic mice on day 10. These two different CD8 T cell populations were used in killing assays. Data are mean ± s.e.m.; an unpaired two-sided t-test was used. H Coculture of ST2⁺ T_reg cells with CD45RA-sorted naïve CD8 T cells, TME-derived CD8 T cells, and CD3/CD8 activated CD8 T cells (beads to cell ratio of 1:1) at T_reg/CD8 ratio of 1:1 and comparison of the cytolytic effect of ST2⁺ T_reg cells on these CD8 T cells (n = 4). Non-TME or TME cell populations were sorted from naïve or leukemic mice on day 10 and used in killing assays. Data are mean ± s.e.m.; ANOVA with post-hoc Bonferroni t-test. I In vivo killing assay of TME BM CD8 T cells by TME BM ST2⁺ T_reg cells. Schema and data of an assay to evaluate T_reg cells’ in vivo cytotoxicity of TME and non-TME CD8 T cells. Representative histograms showing the frequencies of donor-derived TME CD8 T cells (CFSE^high) and non-TME CD8 T cells (CFSE^low) in the BM at 24 h post-adaptive transfer with WT naïve ST2⁻ T_reg cells or TME AML ST2⁺ T_reg cells. The red histograms represent the in vivo proportion of non-TME CD8 T cells (CFSE^low) and TME CD8 T cells (CFSE^high) after adoptive transfer at baseline (0 h; equal proportion) and at 24 h (decreased proportion in the population that has been killed). Statistically analyzed lysis percentages of donor-derived TME CD8 T cells (CD45.1⁺CFSE^high) in the malignant BM niches after 24 h of coculture with WT naïve ST2⁻ T_reg cells or TME BM ST2⁺ T_reg cells (n = 6). Data are mean ± s.e.m.; an unpaired two-sided t-test was used.